infong/php-readability
1
0
Fork 0
mirror of https://github.com/j0k3r/php-readability.git synced 2026-09-27 06:26:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Do not remove nofollow links

Browse Source 
Most the time, they can be usefull.
At least, it'll be a link to something unrelated. But we won't lose a link inside the content.

Also, adding some extra space.
This commit is contained in:
Jeremy Benoist
2015-09-22 19:25:57 +02:00
parent 1830dc45d4
commit b77876b30a
3 changed files with 108 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/ReadabilityTest.php
+2 -2
View File
@@ -112,7 +112,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
public function testStandardClean()
{
$readability = new ReadabilityTested('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<a href="#nofollow" rel="nofollow">will be removed</a></div>', 'http://0.0.0.0');
$readability = new ReadabilityTested('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<a href="#nofollow" rel="nofollow">will NOT be removed</a></div>', 'http://0.0.0.0');
$readability->debug = true;
$readability->lightClean = false;
$res = $readability->init();
@@ -123,7 +123,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
$this->assertContains('<div readability=', $readability->getContent()->innerHTML);
$this->assertEmpty($readability->getTitle()->innerHTML);
$this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
$this->assertNotContains('will be removed', $readability->getContent()->innerHTML);
$this->assertContains('will NOT be removed', $readability->getContent()->innerHTML);
$this->assertNotContains('<h2>', $readability->getContent()->innerHTML);
}